System Reliability Analysis And Maintenance Optimization For Rail Transit Train Under Complicated Coupled Interactions

Reliability is the premise and core competitiveness of rail transit train operation,but high density,complex technology,strong coupling and other factors bring about great challenges to the reliability assurance and operational maintenance management of rail transit trains.The concepts,ideas,models and methods of the traditional reliability assessment and maintenance management have been difficult to deal with the multi-component,multi granularity,strong coupling,nonlinear and dynamic uncertainty of such stochastic process,and cannot reveal and analyze the mechanism,change regularity and regulation mechanism of the risk in the application of the complex system of train.Thus,it is urgent to break through the traditional methods and establish a new reliability evaluation and operational maintenance management method of rail transit train system,open up new perspectives,make full use of the accumulated big data of the operational safety status,comprehensively improve the theoretical methods in quantitative,real-time,detailed scale,individual differentiation and system optimization,as well as improve the prevention and response to the risk of train operation.In view of this,this thesis did some works based on the train design data and the field fault detection records and maintenance data,the following research work is carried out in the aspects of train system risk analysis and key component identification,system failure propagation,system reliability evaluation and multi-component system maintenance optimization.(1)In order to improve the defects of traditional FMECA models,a new analysis model and method of key components identification for train system based on cumulative prospect theory,type 2 intuitionistic fuzzy and VIKOR is proposed.Type-2fuzzy VIKOR method can solve the problem of index fusion in FMECA by entropy weight method.In addition,triangular fuzzy number intuitionistic fuzzy number is used to describe the subjective fuzziness in FMECA.In addition,cumulative prospect theory is introduced to deal with the risk sensitivity and decision-making psychological behavior of FMECA experts.By the actual example of the train system and comparison with other FMECA methods,the new FMECA model and method of key components is verified and analyzed.(2)Based on the knowledge of complex network and virus propagation,the theory of risk potential energy field is proposed,and the probability model of fault propagation of train system is constructed to quantitatively characterize the propagation probability between components.On this basis,the fault propagation process of train system can be simulated by the principle of distribution diffusion,all possible fault propagation paths and their occurrence probability are obtained and thereliability state of the components can be also determined.The results show the feasibility and correctness of the fault path generation model and method based on risk potential energy field.At the same time,the method can quantitatively describe the risk propagation probability between components and analyze the whole process of propagation.The accurate identification of fault path is realized,which is conducive to better train operation and maintenance.(3)In order to make up for the deficiency of the existing train system reliability analysis about ignoring the system polymorphism,a new reliability analysis method of train system by improving the d-MC model is proposed.Taking the bogie system as an example,in order to construct the reliability flow network of the bogie system,three functions of the bogie system and their interaction relations are analyzed based on the complex network theory: bearing function,power transmission function and damping function.On this basis,the unnecessary candidate d-MCs and repeated d-MCs are deleted in advance in the minimal cut analysis,which has dealt with the problem of too much unnecessary data in the existing d-MC theory,and greatly improves the efficiency of system reliability calculation.(4)In order to reduce the cost of train operation and improve the availability of train system,an optimization method of multi-component dynamic maintenance strategy is proposed for train system based on opportunity correlation.Based on the minimum reliability requirements of system components and the principle of increasing failure rate and declining service age,a single component maintenance timing mode is established,and discriminant function of maintenance coefficient is also proposed to select its corresponding maintenance mode.Then,the function structure correlation,failure interaction correlation and reliability state correlation are analyzed in the system.On this basis,an opportunity correlation maintenance model of the train is established.Then,take the lowest cost as the optimization objective,the reliability and availability of components as the constraints,and and the optimal maintenance strategy is calculated by the modified particle swarm optimization algorithm based on the inertia weight factor.The correctness and feasibility of the proposed method is verified through the analysis of the actual example.